updated 2012
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia
Q3: For behavioural and psychological symptoms in people with dementia, do following drugs, when compared to placebo/comparator, produce benefits/harm in the specified outcomes? - conventional antipsychotics - atypical antipsychotics - antidepressant - Trazodone Background Worldwide, there are estimated to be 25 million people with dementia, the majority of them in developed countries have Alzheimer’s disease. However, Alzheimer’s disease accounts for 60% whereas vascular dementia accounts for approximately 30% of the prevalence in low and middle income countries (LAMIC). It is a devastating illness that results in a progressive decline in cognitive ability and functional capacity, causes distress to patients, their carers, and families, and has a large societal impact. Behavioural and psychological symptoms of dementia (BPSD – sometimes referred to as the non-cognitive symptoms of dementia) include: disturbances of behaviour, most of which can be subsumed under the headings of aggression and non-aggressive agitation, and psychological symptoms, particularly visual and auditory hallucinations, persecutory delusions, delusional misidentification, depression and anxiety. BPSD are very common with point prevalence estimates ranging between 60 and 80%, and cumulative risk of 90% across the course of the illness. People with dementia are commonly affected by multiple, and recurrent behavioural problems. BPSD make a large and independent contribution to caregiver strain and are a common precipitating factor for institutionalization. Off-label prescribing of antipsychotic drugs has been commonly employed to treat symptoms of aggression, agitation and psychosis in patients with dementia; in most countries, few or no treatments have been given regulatory approval. Atypical antipsychotic drugs (principally risperidone, olanzapine, quetiapine and aripiprazole) have to a large extent superseded the prescription of conventional (first generation) antipsychotic drugs such as haloperidol and thioridazine.
1
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Given the extent of the concern regarding the efficacy and safety of antipsychotic medications for the treatment of BPSD, it seems very unlikely that any more evidence from randomized placebo-controlled trials will become available to guide policy and practice. It seems likely, at least for the time being, that off-label prescribing of these agents to older people with dementia will continue. This is therefore an opportune time to reappraise the entirety of the evidence pertaining to the balance of benefits and risks associated with the use of these medications. There is clearly a need for safer and more effective remedies, and non-antipsychotic drugs, such as trazodone, carbamazepine, valproate, chloral hydrate, and serotonin re-uptake inhibitor (SSRI) antidepressants have all been advocated. There is little empirical evidence published on the effectiveness of these agents for treating challenging behaviours in elderly people with dementia. Although published evidence is inconclusive, physicians have been using trazodone.
Population/Intervention(s)/Comparison/Outcome(s) (PICO) Population:
Individuals with dementias (cause not specified), Alzheimer’s disease, vascular dementia
Interventions:
Conventional (haloperidol, chlorpromazine and thioridazine), Atypical antipsychotic drugs (risperidone, aripiprazole, quetiapine and olanzapine) Antidepressant (trazodone)
Comparison:
Placebo
Outcomes: CONVENTIONAL ANTIPSYCHOTICS Behavioural symptom change Agitation Aggression Psychosis 2
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Caregiver burden Functional Status (activities of daily living) Adverse effects (Extrapyramidal symptoms) Improving "anxious mood" symptom (thioridazine)
ATYPICAL ANTIPSYCHOTICS Behavioural outcome Improving psychosis Mortality Falls/fractures Cerebrovascular adverse events
TRAZODONE Clinical Global Impression Agitation Cognitive functioning Adverse events (Parkinson gait, akathesia, rigidity)
3
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia
List of the systematic reviews identified by the search process INCLUDED IN GRADE TABLES OR FOOTNOTES CONVENTIONAL ANTIPSYCHOTICS Haloperidol Lonergan E, Luxenberg J, Colford JM (2002). Haloperidol for agitation in dementia. Cochrane Database of Systematic Reviews, (2):CD002852. Thioridazine Kirchner V, Kelly CA, Harvey RJ (2001). Thioridazine for dementia (Review). Cochrane Database of Systematic Reviews, (3):CD000464. ATYPICAL ANTIPSYCHOTICS Ballard C, Waite J (2006). The effectiveness of atypical antipsychotics for the treatment of aggression and psychosis in Alzheimer's disease. Cochrane Database of Systematic Reviews, (1):CD003476. Schneider LS, Dagerman K, Insel PS (2006). Efficacy and adverse effects of atypical antipsychotics for dementia: meta-analysis of randomized, placebocontrolled trials. American Journal of Geriatric Psychiatry, 14:191-210. Ballard C, Howard R (2006). Neuroleptic drugs in dementia: benefits and harm. Nature Reviews Neuroscience, 7:492-500. Prince M, Castro-Costa E, Banerjee S. A systematic review of the evidence on the efficacy and harm associated with the use of conventional and atypical antipsychotics for the treatment of behavioural disturbance and psychosis in dementia (unpublished 2009). TRAZODONE Martinon-Torres G, Fioravanti M, Grimley Evans J (2004). Trazodone for Agitation in dementia. Cochrane Database of Systematic Reviews, 3:CD004990 (last assessed as up to date April 2008).
EXCLUDED FROM GRADE TABLES AND FOOTNOTES (please refer to PICO table) 4
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Ballard C, Waite J (2006). The effectiveness of atypical antipsychotics for the treatment of aggression and psychosis in Alzheimer's disease. Cochrane Database of Systematic Reviews, (1):CD003476. Schneider LS, Dagerman K, Insel PS (2006). Efficacy and adverse effects of atypical antipsychotics for dementia: meta-analysis of randomized, placebocontrolled trials. American Journal of Geriatric Psychiatry, 14(3):191-210. Ballard C, Howard R (2006). Neuroleptic drugs in dementia: benefits and harm. Nature Reviews Neuroscience, 7(6):492-500.
PICO table
Serial no. 1
2
3
Intervention/Comparison Outcomes Haloperidol vs. placebo Agitation Aggression Psychosis Caregiver burden Functional Status (activities of daily living) Adverse effects (extrapyramidal symptoms) Thioridazine vs. placebo Improving psychosis Improving depressive symptoms Improving anxiety symptoms Adverse effects
Systematic reviews used for GRADE Lonergan E, Luxenberg J, Colford JM (2002). Haloperidol for agitation in dementia. Cochrane Database of Systematic Reviews, (2):CD002852.
Explanation This systematic review is the latest review of the area.
Kirchner V, Kelly CA, Harvey RJ (2001). Thioridazine for dementia. Cochrane Database of Systematic Reviews, (3):CD000464.
This systematic review is the latest review of the area.
Atypical antiphychotics vs. placebo
Prince M, Castro-Costa E, Banerjee S. A systematic review of the evidence on the efficacy and harm associated with the use of conventional and atypical antipsychotics for the treatment of
This systematic review considered the evidence for both conventional and atypical antipsychotics vs. placebo; latest review. The unpublished review was
Psychosis Mortality Falls/Fractures Cerebrovascular adverse events
5
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia
4
Trazodone vs. placebo
Clinical Global Impression Agitation Cognitive functioning Adverse event(Parkinson gait, akathesia, rigidity)
behavioural disturbance and psychosis in dementia (unpublished, 2009)
procured from the authors.
Martinon-Torres G, Fioravanti M, Grimley Evans J (2004). Trazodone for Agitation in dementia. Cochrane Database of Systematic Reviews, (3):CD004990 (last assessed as up to date April 2008).
Latest review on the topic
Narrative description of the studies that went into the analysis Haloperidol The review carried out by Lonergan et al, 2002 included five trials. Three studies were multicentre; two studies were from Europe; three studies were from the United States, three studies involved outpatients; two studies involved institutionalized patients. The average age of participants ranged between 72.1 years and 81.5 years; women made up 56.3% to 66.6% of the patients studied. Identification of dementia: all studies used one or more standard methods to diagnose dementia, including Alzheimer’s dementia. Three studies were limited to Alzheimer’s dementia, and two studies included other forms of dementia as well. Characterization of agitation and outcome measures: three studies used the Cohen-Mansfield Agitation Inventory (CMAI) to assess agitation and response to therapy; two studies used the Multidimensional Observation Scale for Elderly Subjects (MOSES); two studies used Brief Psychiatry Rating Scale (BPRS) and one study used the Agitated Behaviour Inventory for Dementia (ABID). Haloperidol dosage varied across the studies from 0.25mg/day to 6.0mg/day. Thioridazine The review carried out by Kirchner et al, 2001 found twelve trials. Four of these had suitable designs but the information reported was too brief to allow analysis. The most common reason for excluding trials was because of an open design. The eight trials that were analysed were similar in design. They were all double-blind, randomized, controlled trials of 3 to 8 weeks duration. Sample size ranged from 30 to 610 participants. Six trials not specify the type of dementia, one specified Alzheimer's Disease and one Vascular dementia. Three trials compared thioridazine with diazepam and one of these had a placebo group. Of the 6
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia remaining trials one compared thioridazine with loxapine and placebo, one with loxapine, one with zuclopentixol, one with etoperidone and one with chlormetiazole . The use of different controls constrained the pooling of results. In comparing thioridazine to diazepam and thioridazine to loxapine some items could be pooled. Mostly data were inadequately reported making pooled analyses impossible. We show only the comparisons of thioridazine with placebo. Chlorpromazine Studies have shown that older adults with dementia who take antipsychotics (medications for mental illness) such as chlorpromazine have an increased chance of death during treatment. Chlorpromazine is not approved by the Food and Drug Administration (FDA) for the treatment of behaviour problems in older adults with dementia. FDA in 2008 notified healthcare professionals that both conventional and atypical antipsychotics are associated with an increased risk of mortality in elderly patients treated for dementia-related psychosis. Atypical antipsychotics The review carried out by Prince et al, 2009 (unpublished) included seventeen trials. For the purposes of its updated meta-analytical review, it focused mainly on the Schneider meta-analytical comparisons (Schneider et al, 2006), which were then extended to include new data from the CATIE-AD trial (two trials). All meta-analyses of individualized outcomes were re-run to generate standardized mean differences, facilitating comparison between drugs and across outcomes, The proportion responding in drug and placebo groups for dichotomized efficacy outcomes were also summarized with absolute risk differences, relative risks rather than odds ratios (which tend to overestimate the relative risk), and numbers needed to treat/ harm with 95% confidence intervals. Trazodone The review carried out by Martinon-Torres et al, 2004 included only two studies comprising 104 participants with dementia. The trials differed in design: one a parallel-group study of patients with Alzheimer’s disease and another a cross-over study of patients with front temporal dementia with an-open label follow-up trial if three years. The results from this extension study have not been used in the analysis. It was not possible to pool the data. The studies were respectively of 16 and six weeks duration, using trazodone from 50 to 300mg daily. Both trials examined global clinical state, behavioural disturbances and cognitive function. Compared with placebo, the use of trazodone was not associated with statistically significant benefits for behavioural manifestations as measured by various rating scales. Analysis of changes from baseline for clinical impression of change and for cognitive function did not produce statistically significant results in favour of trazodone. There is insufficient evidence to recommend the use of trazodone as a treatment for behavioural and psychological manifestations of dementia.
GRADE tables 7
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Table 1 Haloperidol Author(s): Castro-Costa E, Dua Tarun, Huynh N Date: 2009-08-10 Question: Should haloperidol vs. placebo be used for BPSD? Settings: outpatients; inpatients Bibliography: Lonergan E, Luxenberg J, Colford JM (2002). Haloperidol for agitation in dementia. Cochrane Database of Systematic Reviews, (2):CD002852. Summary of findings Quality assessment No of patients No of studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Other considerations
Effect
Importance Quality
haloperidol
placebo
Relative (95% CI)
Absolute
194
175
-
SMD 0.19 lower (0.4 lower to 0.01 higher)
LOW
194
175
-
SMD 0.12 lower (0.33 lower to 0.08 higher)
LOW
254
235
-
SMD 0.31 lower (0.44 to 0.13 lower)
LOW
80/101 (79.2%)
71/103 (68.9%)
OR 1.50 (0.88 to 2.55)
80 more per 1000 (from 28 fewer to 160 more)
VERY LOW
Behavioural symptom change (Better indicated by lower values) 41
randomized trials
serious2
no serious inconsistency
serious3,4
no serious inconsistency
serious3,4
no serious imprecision
none
no serious imprecision
none
no serious imprecision
none
no serious imprecision
none
CRITICAL
Agitation (Better indicated by lower values) 41
randomized trials
serious2
CRITICAL
Aggression (Better indicated by lower values) 31
randomized trials
serious2
no serious inconsistency
serious5
no serious inconsistency
serious5,7
CRITICAL
CGIC(improvement) 21
randomized trials
very serious6
IMPORTANT
caregiver burden (Better indicated by lower values)
8
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 1
randomized trials
very serious8
no serious inconsistency
serious9
serious10
none
serious9
serious10
none
serious3,4
no serious imprecision
none
34
36
-
SMD 0.81 higher (0.89 lower to 2.51 higher) VERY LOW
CRITICAL
34
36
-
SMD 0.91 higher (0.59 lower to 2.41 higher) VERY LOW
CRITICAL
73/295 (24.7%)
72/278 (25.9%)
OR 1.00 (0.68 to 1.46)
0 fewer per 1000 (from 67 fewer to 79 more)
23/135 (17%)
10/139 (7.2%)
OR 2.52 (1.22 to 5.21)
91 more per 1000 (from 14 CRITICAL more to 216 more) MODERATE
34/101 (33.7%)
18/103 (17.5%)
OR 2.34 (1.25 to 4.38)
157 more per 1000 (from 35 more to 306 more)
Functional Status (activities of daily living) (Better indicated by lower values) 1
randomized trials
very serious8
no serious inconsistency
Treatment acceptability(dropouts by endpoint) 4
randomized trials
no serious limitations11
no serious inconsistency
MODERATE
CRITICAL
treatment acceptability(dropouts adverse events) 2
randomized trials
no serious limitations11
no serious inconsistency
serious5,7
no serious imprecision
none
very serious8
no serious inconsistency
no serious indirectness
no serious imprecision
none
Extrapyramidal symptom 1
randomized trials
LOW
CRITICAL
1
From Lonergan et al, 2002 drop outs more than 30% in two study. 3 1 study included all dementia and 3 studies included just Alzheimer’s disease. 4 different scales used to assess behavioural symptoms in 4 trials. 5 One study included all dementia other study included Alzheimer’s disease. 6 drop outs more than 30% in both the included studies. 7 One study is in outpatients while the other study is in hospitalized or nursing home patients. 8 drop out more than 30% in the study. 9 only one study contributes to the evidence base. 10 small sample (less than 100 participants). 11 drop outs were not considered. 2
Table 2 Thioridazine
9
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Author(s): Castro-Costa E, Dua T, Huynh N Date: 2009-08-11 Question: Should thioridazine vs. placebo be used for dementia? Settings: Bibliography: Kirchner V, Kelly CA, Harvey RJ (2001). Thioridazine for dementia. Cochrane Database of Systematic Reviews, (3):CD000464. Summary of findings Quality assessment No of patients No of studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Other considerations
Effect
Importance Quality
thioridazine
placebo
Relative (95% CI)
Absolute
11/17 (64.7%)
14/17 (82.4%)
OR 0.41 (0.09 to 1.86)
167 fewer per 1000 (from 528 fewer to 73 more)
adverse effects 11
randomized trials
very serious2
no serious inconsistency
serious3
very serious4
none
VERY LOW
CRITICAL
VERY LOW
IMPORTANT
VERY LOW
IMPORTANT
VERY LOW
IMPORTANT
Clinical Global evaluation (improvement) 11
randomized trials
serious2
no serious inconsistency
serious3
very serious4
none 10/17 (58.8%) 8/17 (47.1%)
OR 1.58 (0.42 to 114 more per 1000 (from 199 fewer 5.96) to 371 more)
Symptoms: agitation; depressed mood; intellect, behaviour at interview 11
randomized trials
very serious5
no serious inconsistency
serious3
no serious imprecision
none
no serious imprecision
none
135/183 (73.8%)
74/175 (42.3%)
OR 3.64 (2.39 to 304 more per 1000 (from 214 more 5.54) to 379 more)
145/183 (79.2%)
72/175 (41.1%)
OR 4.91 (3.21 to 363 more per 1000 (from 280 more 7.5) to 428 more)
Symptoms: anxious mood; tension; fears; insomnia 11
randomized trials
very serious5
no serious inconsistency
serious3
1
From Analysis of Kirchner et al, 2001 . more than 30% dropouts. 3 only one study contributes to the evidence base. 2
10
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 4 5
sample very small (less than 50 participants). drop outs not discussed.
Table 3 Aripiprazole Author(s): Castro-Costa E, Dua T, Huynh N Date: 2009-08-12 Question: Should aripiprazole vs. placebo be used for treatment of behavioural disturbance and psychosis in Dementia? Settings: Bibliography: Prince M, Castro-Costa E, Banerjee S. A systematic review of the evidence on the efficacy and harm associated with the use of conventional and atypical antipsychotics for the treatment of behavioural disturbance and psychosis in dementia (unpublished, 2009)
Summary of findings Quality assessment No of patients No of studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Other considerations
Effect
Importance Quality
aripiprazole
placebo
Relative (95% CI)
562
324
-
SMD 0.22 lower (0.36 to 0.08 CRITICAL lower) MODERATE
588
388
-
SMD 0.17 lower (0.31 to 0.03 CRITICAL lower) MODERATE
475
236
-
SMD 0.22 lower (0.38 to 0.06 lower) MODERATE
Absolute
BPRStotal (Better indicated by lower values) 3
randomized trials
no serious limitations
no serious inconsistency
serious1,2
no serious imprecision
none
no serious inconsistency
serious1,2
no serious imprecision
none
NPI total (Better indicated by lower values) 33
randomized trials
no serious limitations
Cohen-Mansfield Agitation Inventory (CMAI) total (Better indicated by lower values) 33
randomized trials
no serious limitations
no serious inconsistency
serious1,2
no serious imprecision
none
NPI Psychoses(>50% improvement)
11
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 33
randomized trials
no serious limitations
no serious inconsistency
serious1,2
no serious imprecision
none
randomized trials
no serious limitations
no serious inconsistency
serious1,2
no serious imprecision4
none
no serious limitations
no serious inconsistency
serious1,2
serious4
none
285/588 (48.5%)
129/338 (38.2%)
RR 1.13 (1.01 to 1.27)
50 more per 1000 (from 4 more to 103 more)
MODERATE IMPORTANT
Mortality 33
21/603 (3.5%) 6/348 (1.7%)
RR 1.50 (0.58 to 9 more per 1000 (from 7 fewer IMPORTANT 3.86) to 49 more) MODERATE
8/596 (1.3%) 2/343 (0.6%)
RR 1.29 (0.23 to 2 more per 1000 (from 4 fewer 7.06) to 35 more)
Cerebro vascular event 33
randomized trials
LOW
IMPORTANT
1
2 studies carried out in nursing home patients and 1 study in outpatients. in 2 studies doses are 2-10 mg and in one study 2-15 mg. 3 From analysis of Prince et al, 2009. 4 no effect with upper confidence limit crossing a risk of 2.0. 2
Table 4 Olanzapine Author(s): Castro-Costa E, Dua T, Huynh N Date: 2009-08-12 Question: Should olanzapine vs. placebo be used for treatment of behavioural disturbance and psychosis in dementia? Settings: Bibliography: Prince M, Castro-Costa E, Banerjee S. A systematic review of the evidence on the efficacy and harm associated with the use of conventional and atypical antipsychotics for the treatment of behavioural disturbance and psychosis in dementia (unpublished, 2009)
Summary of findings Quality assessment No of patients No of studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Other considerations
olanzapine
placebo
Effect Relative (95% CI)
Importance Quality Absolute
BPRS incl CATIE-AD 12 weeks (Better indicated by lower values)
12
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 31
randomized trials
no serious limitations
no serious inconsistency
serious2,3
no serious imprecision
none
serious2,3
no serious imprecision
none
serious2,3
no serious imprecision
none
no serious inconsistency
serious2,3
no serious imprecision
none
no serious imprecision
none
serious4
none
808
290
-
SMD 0.13 lower (0.27 lower to CRITICAL 0.01 higher) MODERATE
865
372
-
SMD 0.08 lower (0.2 lower to MODERATE CRITICAL 0.05 higher)
901
313
-
SMD 0.11 lower (0.24 lower to CRITICAL 0.02 higher) MODERATE
958
395
-
SMD 0.10 lower (0.23 lower to CRITICAL 0.03 higher) MODERATE
126/204 (61.8%)
62/94 (66%)
RR 0.94 (0.78 to 1.12)
32/1284 (2.5%)
9/620 (1.5%)
RR 1.48 (0.65 to 7 more per 1000 (from 5 fewer 3.39) to 35 more)
LOW
57/726 (7.9%)
13/223 (5.8%)
RR 1.44 (0.8 to 2.58)
LOW
15/1178 (1.3%)
2/478 (0.4%)
RR 1.81 (0.52 to 3 more per 1000 (from 2 fewer 6.29) to 22 more)
BPRS incl CATIE-AD last phase 1 (Better indicated by lower values) 41
randomized trials
no serious limitations
no serious inconsistency
NPI incl CATIE-AD 12 weeks (Better indicated by lower values) 41
randomized trials
no serious limitations
no serious inconsistency
NPI incl CATIE-AD last phase (Better indicated by lower values) 41
randomized trials
no serious limitations
NPI Psychosis(> 30% improvement) 11
randomized trials
no serious limitations
no serious inconsistency
serious2,3
randomized trials
no serious limitations
no serious inconsistency
serious2,3
randomized trials
no serious limitations
no serious inconsistency
serious3
no serious limitations
no serious inconsistency
serious2,3
40 fewer per 1000 (from 145 fewer to 79 more)
MODERATE
IMPORTANT
Mortality 61
IMPORTANT
Falls 21
serious4
none
26 more per 1000 (from 12 fewer to 92 more)
IMPORTANT
cerebro vascular event 4
1 2
randomized trials
serious4
none
LOW
IMPORTANT
From analysis of Prince et al, 2009 (unpublished). 2 studies carried out in nursing home and 2 studies carried in outpatients.
13
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 3 4
all four studies use different doses. no effect with upper confidence limit crossing risk of 2.
Table 5 Quetiapine Author(s): Castro-Costa E, Dua T, Huynh N Date: 2009-08-12 Question: Should quetiapine vs. placebo be used for the treatment of behavioural disturbance and psychosis in dementia? Settings: Bibliography: Prince M, Castro-Costa E, Banerjee S. A systematic review of the evidence on the efficacy and harm associated with the use of conventional and atypical antipsychotics for the treatment of behavioural disturbance and psychosis in dementia (unpublished, 2009).
Summary of findings Quality assessment No of patients No of studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
Other considerations
Effect
Importance Quality
quetiapine
placebo
Relative (95% CI)
Absolute
154
173
-
SMD 0.25 lower (0.47 to 0.03 lower)
208
155
-
SMD 0.18 lower (0.37 lower to 0 CRITICAL higher) MODERATE
241
92
-
SMD 0.19 lower (0.39 lower to CRITICAL 0.66 higher) MODERATE
BPRS incl CATIE-AD 12 weeks (Better indicated by lower values) 21
randomized trials
no serious limitations
no serious inconsistency
serious2
no serious imprecision
none
MODERATE
CRITICAL
BPRS CATIE-AD incl last phase (Better indicated by lower values) 21
randomized trials
no serious limitations
no serious inconsistency
serious2
no serious imprecision
none
no serious inconsistency
serious2,3
no serious imprecision
none
PANNS-EC (Better indicated by lower values) 11
randomized trials
no serious limitations
CMAI total (Better indicated by lower values)
14
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 11
randomized trials
no serious limitations
no serious inconsistency
serious2,3
no serious inconsistency
serious2,3
no serious inconsistency
serious2,3
serious4
none
27
29
-
SMD 0.13 higher (0.39 lower to 66 higher)
LOW
32/85 (37.6%)
27/94 (28.7%)
RR 1.31 (0.86 to 1.99)
89 more per 1000 (from 40 fewer to 284 more)
MODERATE
99/206 (48.1%)
26/86 (30.2%)
RR 1.59 (1.12 to 2.26)
178 more per 1000 (from 36 more to 381 more)
LOW
75/206 (36.4%)
24/86 (27.9%)
RR 1.30 (0.89 to 1.92)
84 more per 1000 (from 31 fewer to 257 more)
MODERATE
24/485 (4.9%)
10/388 (2.6%)
RR 1.57 (0.73 to 15 more per 1000 (from 7 fewer 3.36) to 61 more)
CRITICAL
BPRS total (>30% improvement) 11
randomized trials
no serious limitations
no serious imprecision
none
serious5
none
IMPORTANT
CGI-C(much;very much improved) 11
randomized trials
no serious limitations
IMPORTANT
PANSS-EC(> 40% improved) 11
randomized trials
no serious limitations
no serious inconsistency
serious2,3
randomized trials
no serious limitations
no serious inconsistency
serious2
no serious imprecision
none
serious5
none
IMPORTANT
Mortality 41
LOW
IMPORTANT
1
from analysis of Prince et al, 2009 (unpublished, 2009). studies with different doses. 3 only one study contributes to evidence base. 4 small sample ( 30% improvement) 11
randomized trials
no serious limitations
no serious inconsistency
serious3,4
no serious imprecision
none
125/196 (63.8%)
62/94 (66%)
randomized trials
no serious limitations
no serious inconsistency
serious3
serious8
none
46/1260 (3.7%)
25/921 (2.7%)
RR 1.29 (0.75 to 8 more per 1000 (from 7 fewer 2.21) to 33 more)
randomized trials
no serious limitations
no serious inconsistency
serious3
no serious imprecision
none
193/1060 (18.2%)
115/665 (17.3%)
RR 1.02 (0.82 to 3 more per 1000 (from 31 fewer IMPORTANT 1.27) to 47 more) MODERATE
no serious limitations
serious9
serious3
serious8
none
37/1175 (3.1%)
7/779 (0.9%)
RR 3.31 (1.58 to 21 more per 1000 (from 5 more IMPORTANT 6.96) to 54 more) VERY LOW
660 fewer per 1000 (from 660 IMPORTANT fewer to 660 fewer) MODERATE
Mortality 61
LOW
IMPORTANT
Falls 41
Cerebrovascular event 71
randomized trials
1
from analysis of Prince et al, 2009 (unpublished). heterogeneity estimate (I sq)= 70%. 3 studies with different doses. 4 only one study contributes to the evidence base. 5 heterogeneity estimate (I sq)= 75.2%. 6 heterogeneity estimate (I sq)=86.8%. 2
17
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia 7
heterogeneity estimate (I sq)=81.9%. no effect with upper confidence limit crossing a risk of 2.0. 9 heterogeneity estimate (I sq)= 62.1%. 8
Table 7 Trazodone Author(s): Castro-Costa E, Dua T, Huynh N Date: 2009-08-29 Question: Should trazodone vs. placebo be used for agitation? Settings: demented patients Bibliography: Martinon-Torres G, Fioravanti M, Grimley Evans J (2004). Trazodone for Agitation in dementia. Cochrane Database of Systematic Reviews, 3:CD004990 (last assessed as up to date April 2008). Summary of findings Quality assessment No of patients No of studies
Design
Limitations
Inconsistency
Indirectness Imprecision
Other considerations
Effect Relative (95% CI)
Importance Quality
trazodone
placebo
Absolute
17/111 (15.3%)
15/108 (13.9%)
74
72
-
MD 0.81 lower (5.07 lower to 3.45 higher)
37
36
-
MD 5.18 higher (2.86 lower to 13.22 higher)
Clinical Global Impression 9improvement) 1
randomized trials
serious1
no serious inconsistency
serious2
serious
none
OR 1.00 (0.57 to 0 fewer per 1000 (from 55 fewer to 83 1.77) more)
VERY LOW
CRITICAL
VERY LOW
CRITICAL
VERY LOW
CRITICAL
Agitation(ABID) (Better indicated by lower values) 1
randomized trials
serious1
no serious inconsistency
serious2
serious3
none
Agitation(CMAI) (Better indicated by lower values) 1
randomized trials
serious1
no serious inconsistency
serious
serious3
none
18
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Memory(RMBPC) (Better indicated by lower values) 1
randomized trials
serious1
no serious inconsistency
serious2
serious3
none
serious3
none
37
36
-
MD 0.11 lower (0.38 lower to 0.16 higher)
37
36
-
MD 1.69 lower (3.18 to 0.2 lower)
74
72
-
MD 0.53 higher (0.39 lower to 1.45 higher)
4/37 (10.8%) 3/36 (8.3%)
OR 1.33 (0.28 to 6.43)
25 more per 1000 (from 59 fewer to 286 more)
1/37 (2.7%) 4/36 (11.1%)
OR 0.22 (0.02 to 2.09)
84 fewer per 1000 (from 109 fewer to 96 more)
3/37 (8.1%) 5/36 (13.9%)
OR 0.55 (0.12 to 2.48)
57 fewer per 1000 (from 120 fewer to 147 more)
IMPORTANT
Cognitive functioning(MMSE) (Better indicated by lower values) 1
randomized trials
serious1
no serious inconsistency
serious2
VERY LOW
IMPORTANT
VERY LOW
IMPORTANT
VERY LOW
CRITICAL
VERY LOW
CRITICAL
VERY LOW
CRITICAL
Activities daily living (Better indicated by lower values) 1
randomized trials
serious1
no serious inconsistency
serious2
no serious inconsistency
serious2
no serious inconsistency
serious2
no serious inconsistency
serious2
serious3
none
Parkinson gait (adverse event 1
randomized trials
serious1
serious3
none
Akathesia(adverse event) 1
randomized trials
serious1
serious3
none
Rigidity(adverse event) 1
randomized trials
serious1
serious3
none
1
dropout not reported. only one study. 3 less than 200 participants. 2
19
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia
Additional information that was not GRADEd Thioridazine In December 2000, the Committee for Safety of Medicines (CSM), UK, issued an urgent cascade fax to all GPs (Wright et al, 2004), recommending that thioridazine should be restricted to second-line treatment of schizophrenia in adults and that the balance of risks and benefits is unfavourable for its previous indications (anxiety, agitation and restlessness in the elderly, moderate to severe psychomotor retardation, violent and dangerously impulsive behaviour, mania/hypomania, and behavioural disorders and epilepsy in children). The fax recommended that treatment with thioridazine should be supervised by a consultant psychiatrist (with monitoring of the patient's electroencephalogram [ECG] and electrolytes), because of concerns that thioridazine was associated with a higher risk of life-threatening arrhythmias and sudden death. Haloperidol Mortality has not been reported as an outcome in randomized controlled trials of haloperidol in dementia. There is insufficient evidence from randomized controlled trials to establish an increased mortality risk associated with the use of haloperidol and other conventional antipsychotics in dementia. However, data from observational studies that either compare all older users of atypical and conventional antipsychotics, or restrict these comparisons to older people with dementia, consistently suggest a slightly higher mortality (around 30%) among users of conventional antipsychotics, similar to the mortality effect observed for atypical antipsychotics. Atypical antipsychotics In October 2002 the manufacturer of risperidone notified Canadian healthcare professionals that in drug-sponsored clinical trials, risperidone users had been found to have a higher rate of cerebrovascular adverse events (CVAEs) relative to those receiving placebo. In 2003, the Food and Drug Administration (FDA) and other authorities published warnings and required changes to the prescribing information for risperidone. At the beginning of 2004, the European Agency for the evaluation of Medicinal Products (EMEA) also issued public advice about an increased risk of CVAEs and mortality in elderly patients with dementia receiving olanzapine. In March 2004, the UK Medicines and Healthcare products Regulatory Agency (MHRA) informed clinicians that risperidone and olanzapine should no be used to treat behavioural and psychological symptoms of dementia because of increased risk of strokes with both drugs . Similarly, in 2005 the FDA issued warnings for aripiprazole regarding the risk of CVAEs, including stroke, in elderly patients with dementia. In April 2005 the FDA informed health professionals of the results of an independent, pooled analysis of 17 RCTs reporting a 1.7 times increased risk of all cause mortality associated with atypical antipsychotic use.
20
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia
Reference List Ballard C, Howard R (2006). Neuroleptic drugs in dementia: benefits and harm. Nature Reviews Neuroscience, 7:492-500. Ballard C, Waite J (2006). The effectiveness of atypical antipsychotics for the treatment of aggression and psychosis in Alzheimer's disease. Cochrane Database of Systematic Reviews, (1):CD003476. FDA (2008). Information for Healthcare Professionals: Conventional Antipsychotics. (http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm124830.htm, accessed 15 February 2002). Kirchner V, Kelly CA, Harvey RJ (2001). Thioridazine for dementia (Review). Cochrane Database of Systematic Reviews, (3):CD000464. Lonergan E, Luxenberg J, Colford JM (2002). Haloperidol for agitation in dementia. Cochrane Database of Systematic Reviews, (2):CD002852. Martinon-Torres G, Fioravanti M, Grimley Evans J (2004). Trazodone for Agitation in dementia. Cochrane Database of Systematic Reviews, 3:CD004990 (last assessed as up to date April 2008). Prince M, Castro-Costa E, Banerjee S. A systematic review of the evidence on the efficacy and harm associated with the use of conventional and atypical antipsychotics for the treatment of behavioural disturbance and psychosis in dementia (unpublished 2009). Schneider LS, Dagerman K, Insel PS (2006). Efficacy and adverse effects of atypical antipsychotics for dementia: meta-analysis of randomized, placebocontrolled trials. American Journal of Geriatric Psychiatry, 14:191-210. Wright NMJ et al (2004). Impact of the CSM advice on thioridazine on general practitioner prescribing behaviour in Leeds: time series analysis. British Journal of Clinical Practice, 54:370-3.
From evidence to recommendations
Factor Narrative summary of the
Explanation Haloperidol
21
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia evidence base
Summary of the quality of evidence
Behavioural symptoms SSssSSsssss 4 studies, SMD -0.19(-0.40 to 0.01, no difference) change Agitation
4 studies, SMD -0.12(-0.33 to 0.08, no difference)
Aggression
3 studies, SMD -0.31(-0.44 to -0.13 favouring active treatment)
Improvement Psychosis
2 studies, OR 1.50(-0.88 to 2.55, no difference)
Caregiver burden
1 study, SMD 0.81(-0.89 to 2.51 no difference)
Functional status (activities of daily living)
1 study, SMD 0.91(-0.59 to 2.41 no difference)
Treatment acceptability (dropouts by endpoint)
4 studies, OR 1.00(-0.68 to 1.46 no difference)
Treatment acceptability (dropouts by adverse events)
2 studies, OR 2.52(1.22 to 4.38 favouring active treatment)
Extrapyramidal symptoms
1 study, OR 2.34(1.25 to 4.38 favouring active treatment)
Behavioural symptoms change
LOW
Agitation
LOW
Aggression
LOW
Improvement Psychosis
VERY LOW
Caregiver burden
VERY LOW
Functional status (activities of daily living)
VERY LOW
22
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Treatment acceptability (dropouts by endpoint)
MODERATE
Treatment acceptability (dropouts by adverse events)
MODERATE
Extrapyramidal symptom
LOW
Balance of benefits versus harms
Other than aggression, no other outcome assessing aspects of behavioural disturbance and psychosis in dementia is improved by use of haloperidol. Additionally, haloperidol could not be shown to reduce caregiver burden or to improve functional status among treated patients, compared with controls. Finally, treatment with haloperidol is associated with higher rates of adverse events such as extrapyramidal symptoms. There is insufficient evidence from randomized controlled trials to establish an increased mortality risk associated with the use of haloperidol and other conventional antipsychotics in dementia. However, data from observational studies that either compare all older users of atypical and conventional antipsychotics, or restrict these comparisons to older people with dementia, consistently suggest a slightly higher mortality (around 30%) among users of conventional antipsychotics.
Values and preferences including any variability and human rights issues
Behavioural disturbance and psychosis associated with the cognitive decline of dementia represents a serious burden for patients and their families, particularly severe and uncontrolled agitation and aggression, where both the person with dementia, and carers may be placed at risk of harm.. However, inefficacy and adverse effects represent serious concerns regarding the use of haloperidol.
Costs and resource use and any other relevant feasibility issues
Included in WHO Essential Medicines list
Narrative summary of the evidence base
Thioridazine Adverse effects
SSssSSsss
1 study, OR 0.41 (0.09 to 1.86, no difference)
Improving psychoses (clinical global evaluation)
1 study, OR 1.58 (0.42 to 5.96, no difference)
23
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia
Summary of the quality of evidence
Balance of benefits versus harms
Improving depressive symptoms
1 study, OR 3.64 (2.39 to 5.54, favouring active treatment)
Improving anxious symptoms(anxious mood; tension; fears; insomnia)
1 study, OR 4.91 (3.21 to 7.5, favouring active treatment)
Adverse effects
VERY LOW
Improving psychoses (clinical global evaluation
VERY LOW
Improving anxious symptoms(intellect, agitation; depressed mood; behaviour at interview
VERY LOW
Adverse effects
VERY LOW
The evidence is inconclusive and so it is not possible to determine if there is a clinically significant difference between thioridazine and placebo in people with dementia. The only positive effect of thioridazine when compared to placebo is to reduce anxiety. It has high rates of adverse effects and no effect on global ratings. Serious cardiotoxicity has been detected on routine monitoring.
Values and preferences including any variability and human rights issues
Behavioural disturbance and psychosis associated with the cognitive decline of dementia represents a serious burden for patients and their families. However, inefficacy and adverse effects represent serious concerns regarding the use of thioridazine
Costs and resource use and any other relevant feasibility issues
24
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Atypical antipsychotics Narrative summary of the evidence base
Outcome
Aripiprazole
Olanzapine
Quetiapine
Risperidone
Behavioural outcomes BPRS total
3 studies, SMD -0.22(0.36 to -0.08, favouring active treatment)
4 studies, SMD -0.08(0.20 to 0.05, no difference)
2 studies, SMD-0.18(0.37 to 0.00, no difference)
2 studies, SMD -0.03(-0.22 to 0.16, no difference)
NPI total
3 studies, SMD -0.17(0.31 to -0.03, favouring active treatment)
4 studies, SMD -0.10(0.23 to 0.03, no difference)
-
2 studies, SMD -0.11(-0.29 to 0.08, no difference)
CMAI total
3 studies, SMD -0.22(0.38 to -0.06, favouring active treatment)
-
1 study, SMD 0.13(0.39 to 0.66, no difference)
PANSS-EC
-
-
1 study, SMD -0.19(0.43 to 0.05, no
3 studies, SMD -0.31(-0.44 to -0.19, favouring active treatment)
-
25
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia difference) BEHAV-AD total
-
-
-
4 studies, SMD -0.19(-0.29 to -0.08, favouring active treatment)
CGI-S total
-
-
-
3 studies, SMD -0.19(-0.32 to -0.06, favouring active treatment)
Improving psychosis NPI Psychosis
3 studies, RR 1.13(1.011.27, favouring active treatment)
1 study, RR 0.94(0.781.12, no difference)
-
BPRS total
-
-
1 study, RR 1.31(0.861.99, no difference)
CGI-C
-
-
1 study, RR 1.59(1.122.26, favouring active treatment
PANSS-EC
-
-
1 study, RR 1.30(0.891.92, no
1 study, RR 0.97(0.81-1.16, no difference)
-
2 studies, RR 1.35(1.19-1.54, favouring active treatment)
-
26
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia difference) Behave-AD
-
-
-
3 studies, RR 1.42(1.21-1.67, favouring active treatment
Adverse events Mortality
3 studies, RR 1.50(0.583.86, no difference)
6 studies, RR 1.48(0.653.39, no difference)
4 studies, RR 1.57(0.733.36, no difference)
-
2 studies, RR 1.44(0.802.58, no difference)
-
3 studies, RR 1.29(0.237.06, no difference)
5 studies, RR 1.81(0.526.29, no difference)
-
Falls
Cerebrovascular event
Summary of the quality of evidence
Outcome
Aripiprazole
6 studies, RR 1.41(0.99-2.03, no difference
-
7 studies, RR 3.31(1.58-6.96, favouring active treatment)
Olanzapine
Quetiapine
Risperidone
MODERATE
MODERATE
Behavioural outcome BPRS total
MODERATE
MODERATE
NPI total
MODERATE
MODERATE
CMAI total
MODERATE
PANSS-EC
-
-
-
LOW
-
MODERATE
LOW MODERATE 27
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia BEHAV-AD total
-
-
-
LOW
CGI-S total
-
-
-
VERY LOW
-
MODERATE
Improving psychosis NPI Psychosis
MODERATE
MODERATE
BPRS total
-
-
MODERATE
CGI-C
-
-
LOW
PANSS-EC
-
-
MODERATE
Behave-AD
-
-
VERY LOW
-
MODERATE
Adverse events Mortality
MODERATE
Falls Cerebrovascular event Balance of benefits versus harms
LOW
LOW
LOW
LOW
LOW
-
MODERATE
LOW
-
VERY LOW
Atypical antipsychotic drugs show minimal efficacy for the treatment of behavioural symptoms in dementia. There is some evidence to support slightly greater efficacy for risperidone for the treatment of aggression, as opposed to non-aggressive agitation. Also, there is very limited evidence indeed for any efficacy of atypical antipsychotic drugs for the treatment of symptoms of psychosis. For a clinically significant degree of improvement, this has only been demonstrated for aripiprazole. Regarding adverse effects, there is suggestive evidence from randomized, placebo-controlled trial, of increased mortality associated with the use of atypical antipsychotic drugs. There is fairly consistent evidence from randomized controlled trials of risperidone and olanzapine that use of these drugs in dementia is not associated with an 28
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia increased incidence of falls, at least over the typical 6-12 week duration of these studies. Meta-analysed evidence from 15 randomized placebo controlled trials of atypical antipsychotics provides robust evidence for an increased risk of cerebrovascular adverse events. Values and preferences including any variability and human rights issues
Behavioural disturbance and psychosis associated with the cognitive decline of dementia represents a serious burden for patients and their families. However, the inefficacy and important adverse effects such as increased mortality and increased cerebrovascular events may represent serious concerns to the use of atypical antipsychotics.
Costs and resource use and any other relevant feasibility issues
Atypical antipsychotics are associated with high acquisition costs, and treatment requires resource use and clinical and laboratory monitoring Atypical antipsychotics are not included in the WHO list of essential medicines.
Trazodone Narrative summary of the evidence base
Clinical Global Impression
1 study, OR 1.00 (0.57 to 1.77, no difference)
Agitation (ABID)
1 study, MD -0.81(-0.507 to 3.45, no difference)
Agitation (CMAI)
1 study, MD 5.18(-2.86 to 13.22, no difference)
Memory (RMBPC)
1 study, MD -0.11(-0.38 to 0.16, no difference)
Cognitive functioning (MMSE)
1 study, MD -1.69(-3.18 to -0.20, favouring active treatment)
Activities of daily living
1 study, MD 0.53(-0.39 to 1.45, no difference)
Parkinson gait
1 study, OR 1.33(0.28 to 6.43, no difference)
Akathesia
1 study, OR 0.22(0.02 to 2.09, no difference)
29
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Rigidity
1 study, OR 0.55(0.12 to 2.48, no difference)
Summary of the quality of evidence
VERY LOW for all the outcomes
Balance of benefits versus harms
The evidence is inconclusive and so it is not possible to determine if there is a clinically significant difference between trazodone and placebo in people with dementia. There is extremely limited data available to support the use of trazodone in the treatment of behavioural disturbance and psychosis in dementia. There is no positive effect of trazodone when compared to placebo to reduce behavioural disturbance and psychosis in dementia.
Values and preferences including any variability and human rights issues
Behavioural and psychiatric disturbances affect at least 50% of people with Alzheimer’s disease and other dementias. Agitation and other challenging manifestations are common, major problems in clinical management. Aberrant behaviours contribute a great deal to caregiver burden and can make home care for patients very difficult. They are generally associated with the need for institutionalisation, rapid cognitive decline, impaired quality of life and increased healthcare costs.
Costs and resource use and any other relevant feasibility issues
Not included in WHO Essential Medicines List
Final recommendation(s) Thioridazine, chlorpromazine or trazodone should not be used for the treatment of behavioural and psychological symptoms of dementia. Strength of recommendation: STRONG Haloperidol and atypical antipsychotics should not be used as first line management for behavioural and psychological symptoms of dementia. Where there is clear and imminent risk of harm with severe and distressing symptoms, the short term use of haloperidol or atypical antipsychotic medications may be considered, preferably with specialist inputs. To the extent possible, informed consent and agreement should be obtained from the person and carer with regard to balance of risk and benefit. 30
Conventional and atypical antipsychotics & antidepressant (trazodone) for behavioural and psychological symptoms in people with dementia Strength of recommendation: STRONG
Limitations The long-term consequences of antipsychotic exposure in this patient population have not been considered in the trials included in this evidence profile. Additionally, it is unclear how different antipsychotics compare in terms of efficacy and tolerability profile.
Update of the literature search – June 2012 In June 2012 the literature search for this scoping question was updated. The following systematic review was found to be relevant without changing the recommendation:
Lonergan E, Luxenberg J, Colford JM, Birks J. Haloperidol for agitation in dementia. Cochrane Database of Systematic Reviews 2002, Issue 2. Art. No.: CD002852. DOI: 10.1002/14651858.CD002852.
31
